Composite lining of deep buried water conveyance tunnel for bearing high internal water pressure is a new type of applicable structure. However, up to date, no effective method is available to calculate the stress of the structure. In this paper, a simplified algorithm, which can be used to calculate the stress distribution of composite lining accurately but costs little computational resource, is proposed. This algorithm, which is based on the elastic theory, takes the effect of internal water pressure and surrounding rock on the composite lining into consideration, respectively. Then, the stress distribution of composite lining in infinite body is derived on the basis of Lame solution. Finally, a case study is followed by choosing a typical section of the Eastern Canal in Beijing of the South‐to‐North Water Diversion Project (SNWDP). This case study was analysed by using the simplified algorithm and verified by finite element method with ABAQUS. The results show that the stress distribution of composite lining can be obtained quickly and accurately with the simplified algorithm, which can provide a reference for other engineering designs.

Traditional methods for geostressfield inversion face issues such as weak physical interpretability and insufficient generalization ability. This study pioneers the application of Physics-Informed Neural Network (PINN) to this problem, developing a data- and physics-driven inversion algorithm. The framework incorporates a constitutive-equation-based regularized loss function as a hard constraint during training to ensure physical consistency. To address boundary load uncertainty, two quantification approaches—Bayesian linear regression and surrogate model optimization—are proposed to establish 95% confidence intervals for boundary coefficients. Verification based on simple three-dimensional models and actual geological models of mines shows that PINN inversion achieves a mean absolute relative error as low as 0.0772%, with an error of 15.67% under sparse sampling conditions—significantly lower than the 31.07% error of the traditional Back propagation neural network. This demonstrates excellent robustness and data efficiency. In the practical engineering application of complex geological bodies, the average error of principal stress inversion is 9.35% with a minimum error of 0.137%. All inversion results fall within the permissible accuracy range of engineering, and the stress distribution conforms to basic laws, with an average error of 0.453 in the constitutive relation. Compared with BP neural network and multiple linear regression methods, it shows obvious accuracy advantages. This method provides a new solution for intelligent ground stress prediction with high accuracy, high efficiency, and strong physical interpretability, and also lays the foundation for early identification of geological disasters.

In situ stress state is a predominant factor for the design and safe construction of geotechnical engineering. For a real construction site, the amount of calculation using a finite element method for in situ stress field increases dramatically with the increase of the calculation freedom due to large‐scale uncertainties. In order to reduce the computing cost without losing the accuracy of the calculation, an optimization algorithm combined with a reduced order model, which is realized by the proper orthogonal decomposition algorithm (POD) for large‐scale in situ stress field, is put forward in this paper. The POD algorithm produces a set of orthogonal bases through the extraction of the field variables, combining with the Galerkin finite element method to create a reduced order numerical model. The reduced order model is then calculated with a global optimization algorithm to inversely find the solution for the actual in situ stress field. In order to verify the accuracy and efficiency of the method, two examples are presented to simulate the inverse calculation of the in situ stress field. They showed that the computation time of the POD method could reach 1/10 of the ordinary computation time. Also, the results showed good accuracy with a minimum computational expense, which can provide a reference for inverse calculation of large‐scale in situ stress field.

For projects such as tailings ponds, slopes, and foundations, loose materials such as rock, slag, and sand, which are composed of particles, often have low cohesion and rely mainly on friction to maintain stability. The shear strength parameters, namely, the internal friction angle and cohesion, are the core parameters that describe the mechanical properties of materials and are directly related to the engineering stability of the above projects. The shear strength properties of loose media are related to the geometric morphological characteristics of particles. Particles with high irregularity will increase the bite and friction of the contact interface between particles, thereby affecting the overall peak shear strength of the material. This study takes sand as the research object. Based on the Mask R-CNN algorithm in deep learning, a sand particle image dataset consisting of single, contact, and sand surface particles is established. An image segmentation model that can identify particles on the surface of the sand layer and obtain the corresponding particle mask is trained; a Python 3.11.4 program is written to automatically calculate seven characteristic parameters of particle morphological characteristics parameters, including the Feret major diameter, the particle Feret minor diameter, the particle aspect ratio, the particle roundness, the comprehensive shape coefficient, the roughness, and the convexity through the particle mask. This method can obtain the overall morphological characteristics of sand particles in real time and is a particle processing method that is a prerequisite for the subsequent rapid prediction of the strength properties of granular materials.